The project is a retrospective analysis of prescribing patterns for patients with hypertension, utilizing two data sources: the 1987 National Medical Expenditure Survey and a database containing claims from a number of self-insured employers in New York state from 1992-96. The study aims to determine how a number of clinical, economic, and sociodemographic factors affect the decision to use medications to treat hypertension and, once the decision to treat is made, how these factors affect patterns of use of antihypertensive medications, particularly in relation to published recommendations for treatment. The analytic approach is an econometric analysis using instrumental variables (IV) and is based on the theoretical model of health capital described by Grossman.